Many people are becoming curious, with the increasing popularity of cloud topology, as to what cloud computing deployment models exist, and which ones are popular. While cloud is a big buzzword right now, a lot of people are kind of mystified in regards to what it really is. That’s ok, that’s what I’m here for.
So, today, I’m going to clarify once more, for those new here, what cloud computing is, then go over some cloud computing deployment models which are popular, and maybe talk a little bit about how they work.
First off, cloud just means that it exists off location, and is being stored, processed and/or served by an outside machine or machines. This is usually utilized so that software or processes can be controlled from devices too weak to actually perform them, but which clusters of servers, or one really strong server, can execute easily. It is an effective way of putting super computing into Joe Everyguy’s hands without a lot of fuss.
It’s also pursued as a safe and secure backup and storage system, as well as used for communications and cooperative software use and work performance online. So it’s pretty much anything web sources or web powered that’s not local network.
So, what are the popular deployment models for cloud computing? In general, there are three of these models that are standardly used, so let’s take a look at what they are.
The simplest model is just a typical webhost scenario, where you use a single dedicated server or host space on a server, and use limited enhanced computing and cloud services from this single source. It is the most affordable and easy to work with.
The second is the grid, which utilizes a series of servers, usually in the same datacenter, linked together to form a great, herculean computing force for all users logged into it. It’s the most expensive model, due to its specificity, but is also not the most powerful one, in technicality, which is …
Cluster computing. Cluster computing maps a global range of available servers regardless of space. It will link them together to form nodes of shared super computing power, often based on distance from the user. If any fail, the next nearest would take its place in an almost cellular or neural configuration. This system is prone to some faults, but is the most dynamic despite being only the middle of the road for pricing.
Algorithms for this model need to be refined and protocols adjusted, but what system doesn’t need that starting out?
If you’re interested in cloud computing deployment models, there exist very long PDFs that go into the deep, scientific details about the different architectures, including less-used ones than the ones I mention here. They also have, gasp, diagrams, which I cannot provide for you.
Cloud computing has some challenges ahead of it in being accepted and made completely practical for its intended range of purposes, but one day, it will take the place of traditional computing in many arenas, so we may as well embrace it now, and understand it to the fullest. Tomorrow waits for nobody, but nobody said we can’t get there early, right?